Christopher Potts, The logic of conventional implicatures (Oxford

Paul Grice warned that 'the nature of conventional implicature needs to be examined before any free use of it, for explanatory purposes, can be indulged in' (1978/1989: 46). Christopher Potts heeds this warning, brilliantly and boldly. Starting with a definition drawn from Grice's few brief remarks on the subject, he distinguishes conventional implicature from other phenomena with which it might be confused, identifies a variety of common but little-studied kinds of expressions that give rise to it, and develops a formal, multidimensional semantic framework for systematically capturing its distinctive character. The book is a virtuosic blend of astute descriptive observations and technically sophisticated formulations. Fortunately for the technically unsophisticated reader, the descriptive observations can be appreciated on their own. Here is a quick summary of the book. Following a brief introductory chapter (so titled), chapter 2 makes 'A preliminary case for conventional implicatures' by offering a four-part definition, distinguishing conventional implicature (CI) from conversational implicature and presupposition, identifying the main kinds of linguistic phenomena that fit this definition, and motivating the book's distinctive multidimensional semantic framework. Chapter 3, 'A logic for conventional implicatures', develops a rigorous 'description logic' for representing CI-meanings along with 'at-issue' meanings. The next two chapters illustrate and discuss the two main kinds of expressions with CI-meanings. Chapter 4 focuses on 'Supplements', including nonrestrictive relative clauses, as-parentheticals, Noun Phrase appositives, and several sorts of adverbials; and chapter 5 considers 'Expressive content', including expressive attributive adjectives, epithets, and Japanese honorifics. Chapter 6, 'The supplement relation: a syntactic alternative', compares Potts's approach to supplements with an alternative syntactic approach, and argues that there is no need to complicate the syntax in order to distinguish supplement structures from coordination – the difference can be captured with multidimensional semantics. The seventh and last chapter, 'A look outside Grice's definition', briefly considers what sorts of linguistic phenomena do, or in principle could, arise when one or another of the four conditions in the definition of conventional implicature is not satisfied. The most important case of this involves words like